Wait, am I Being Fair? Characterizing Deductive Stereotyping and Mitigating It with Fair-GCG

Researchers at arXiv identify a failure mode in large language models called deductive stereotyping, where models apply population-level statistics to individuals, producing biased inferences. They propose Fair-GCG, a reasoning-time injection framework that discovers phrases to steer models toward fairness-aware reasoning, improving performance across fairness benchmarks and reducing bias in open-ended generation.

arXiv:2606.30989v1 Announce Type: new Abstract: Warning: This paper contains several toxic and offensive statements. While reasoning generally improves fairness in recent large language models LLMs , failures persist. In this work, we identify a failure mode, deductive stereotyping, in which models apply population-level statistical regularities to individual cases, producing logically coherent yet socially biased inferences. We provide a statistical interpretation of this phenomenon. To steer models toward fairness-aware reasoning, we propose a reasoning-time injection framework. We further introduce Fair-GCG to systematically discover effective injection phrases. Injection phrases discovered by Fair-GCG improve performance across multiple fairness benchmarks, generalize from smaller to larger LLMs, improves reasoning-level fairness, reduces bias in open-ended generation, and transfer to real-world fairness-sensitive tasks.